This is a technology column, not especially a business column, but just maybe you read the paper or watched television this week and noticed technology stocks entering what appeared to be a free fall. How the heck did that happen and where does it end? That's what this week's column is about. And even if you are the nerdiest of the nerds, getting a better understanding of why you are not quite so rich as you were a week ago is worth doing. So please, read on.

In the 18th century, Adam Smith wrote of an "invisible hand" that guided the economy unerringly in the proper direction. In the 20th century, that hand was recast as an efficient market that had perfect intelligence, always adjusting in advance to news most of us hadn't yet heard. But on the eve of the 21st century, it appears that even market intelligence is having difficulty keeping up with Internet speed. For an example of this just look at 3Com, an outfit that might be killed by success. 3Com is a venerable company in the networking business, founded almost 20 years ago by Bob Metcalfe, the charismatic inventor of Ethernet, still the most popular way of linking computers together. 3Com makes more Ethernet adapter cards than any other company, though it may be more famous lately for its Palm Computing division, recently spun-off through a highly successful initial public offering. Palm Inc. makes the Palm Pilot hand held organizers. These devices, which have been sold in the millions, dominate their niche in defiance of Microsoft's best competitive efforts. 3Com bought Palm a few years ago for not very much money and saw the operation become profitable, then a de facto standard, and finally a Wall Street darling. This ought to be a happy story for 3Com, but in many ways it isn't. From the perspective of 3Com, the Palm IPO probably looks like a disaster. Or does it?

Here is 3Com's apparent problem. Palm Inc. went public on March 2, priced at $38 per share. Following the recent trend of hot IPOs with relatively small numbers of available shares, Palm stock soared on speculative fever and day trading before dropping back below the offering price by April 4 on bad earnings news. Even then, Palm's market capitalization was $21.1 billion to 3Com's $15.8 billion. These numbers make no sense at all, given the fact that 3Com owns more than 95 percent of Palm. Ninety-five point nine percent of $21.1 billion is just over $20 billion, suggesting the perfectly efficient market believes the non-Palm parts of 3Com representing $5.5 billion in sales and hundreds of millions in profit were worth a negative $4.245 billion. That's an implied Price-to-Earnings ratio of about minus 9.

Negative PE ratios aren't supposed to exist. If they did exist, the theory goes, it would be for only a short time before that old invisible hand would go to work readjusting prices and arbitraging away the price differential. Quantitative and program traders would quickly suck any available profits out of the price difference, buying even bigger homes in the Hamptons in exchange for returning the market to stability. Only that's not what happened for 3Com. The crazy price differential continued.

Faced with these numbers, it would be nice if we could just roll our collective eyes and make jokes about that wacky market for tech stocks, except that's not the way things work. There either is an efficient market or there isn't. We can't have it both ways.

What's going on here? Structurally, it looks something like the heyday of leveraged buyouts from the 1980s. Back then the idea was that the parts of many inefficient companies could be sold piecemeal for more than the market capitalization of the company as a whole. It appeared to be good business to take companies private and sell off the under performing parts. But a lot has changed since the 1980s. Industries are leaner and, most importantly, markets are quicker and more efficient. Sure, there might be a little fat inside 3Com, but not that much. Yet this is exactly the explanation that Wall Street has accepted. The conceit of the efficient market has analysts working overtime trying to justify these numbers, citing them as a sign of troubles with 3Com's core networking business. It is as though Palm, itself, has no role in this adventure: The cause is perceived as being all 3Com.

The truth is somewhat different, however. The very success of Palm sent 3Com into a fit of restructuring, the extent of which would be difficult to otherwise justify. On March 20, the company announced plans to sell or abandon product lines representing $3 billion in annual sales, effectively cutting the size of the company in half.

In order to justify the screwy 3Com/Palm stock price differential, securities analysts have pointed to supposed problems with 3Com's cost structure, when the real problem is probably more with the PRICE structures of both stocks. This is the practical difference between a theoretically efficient market and reality. Several factors are inhibiting market efficiency. Certainly 3Com has had its share of troubles in recent years, and that news causes friction that tends to hold down the price of the mother company's shares. This is fairly common among established technology companies that have new (typically Internet) divisions whose real value they feel isn't recognized by the market. The usual answer to this problem is to establish a tracking stock. Yet that's exactly what Palm Computing was for 3Com, and the company was still hurt.

Just as friction may have tended to hold down the price of 3Com shares, speculation and IPO fever made Palm shares excessively buoyant. If we still believe here in market efficiency it must be qualified as in "the market will be EVENTUALLY efficient." The market IS efficient, I believe, and the stock prices eventually WILL realign, and when they do so, it will be a day of sober reckoning. Viewing Palm's stock from the perspective of the more liquid 3Com shares (that is, using 3Com to track Palm, rather than vice versa) suggests a heavily discounted real value for Palm on the order of $10 per share. Wow!

Go back and reread that last paragraph, because it pretty much explains why stocks have fallen this week and how far we might expect them to drop before recovering. And I believe they WILL recover, but that's another story.

The big question, of course, is why we tolerate this price differential in the first place? Does this kind of problem exist in many IPOs involving spinoffs? Most people, including the business media, didn't even notice. Maybe it happens all the time. Certainly, I am not claiming here that any laws were broken.

The price differential wouldn't exist at all in a world populated solely by accountants. But it turns out to be in the interest of many other parties to ignore the problem �- parties including the executive teams of both companies, the entire Wall Street community, and largish populations of speculators, each hoping to be the last out the trading door before the lights are turned off. For the other Palm investors, maybe it's a simple psychological effect; the more something hurts, the more it must be worth doing.

In the end, the upside for Palm the child is greater than the downside for 3Com the parent, a company that probably will be improved by restructuring. In fact, having taken $874 million out of Palm in the first stock sale with more sales to come, 3Com is doing quite well for its apparent humiliation. And Wall Street gains big time through commissions from Palm's success. Certainly it is suspicious that a topic involving so much money is questioned so little.